License
Two licenses apply, because two different things are published here: the software that builds the archive, and the records the archive contains. The records inherit their terms from Wikipedia, so they carry obligations that the code does not.
| What | License | You must |
|---|---|---|
The records — everything under /api/ and every day page |
CC BY-SA 4.0 | Credit this archive and its sources, and license derivatives alike |
| The census readings | CC BY-SA 4.0 | Same. These are our own observations, released on the same terms |
| The software that builds it | MIT | Keep the copyright notice |
Because the archive incorporates material from Wikipedia, the records are published under the Creative Commons Attribution-ShareAlike 4.0 International license. That is not a choice — share-alike is inherited, and passing it on is a condition of using Wikipedia's text at all.
Credit the archive and the underlying source.
Every item in every record carries a link to where it came from. Keep those links when you reuse the data — they are how a reader gets back to the primary source, which is the entire point of an index.
License what you build on it the same way.
If you remix or build upon this material, distribute your contributions under the same license.
A suggested attribution line:
Data from 28081999.me (CC BY-SA 4.0), incorporating material from Wikipedia contributors (CC BY-SA 4.0), Hacker News and arXiv. https://28081999.me
This archive indexes public sources. It does not replace them, and it claims no ownership over their content.
| Layer | Source | Their terms |
|---|---|---|
| attention | Wikipedia pageviews, via the Wikimedia API | CC BY-SA 4.0 |
| history | Wikipedia On this day | CC BY-SA 4.0 |
| frontier | Hacker News | Public API. Titles and links only, always credited and linked back to the discussion |
| research | arXiv | Open metadata under the arXiv API terms. Abstracts are excerpted and linked, never rehosted in full |
| census | Our own measurement of 2,110 public origins | Original observation, released CC BY-SA 4.0 |
| panel | Tranco research ranking | Freely available for research; the list ID used is recorded in every reading |
Wikipedia is a registered trademark of the Wikimedia Foundation. Hacker News is operated by Y Combinator. arXiv is operated by Cornell University. None of them endorse this project.
MIT License Copyright (c) 2026 28081999.me Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Records are immutable by design — this is a record of what was observable on a day, not a tidied retrospective — but immutability is not a license to be wrong.
If something here is factually incorrect, the correction almost always belongs upstream: an item's text comes from Wikipedia, Hacker News or arXiv, and fixing it there is what makes it right everywhere rather than only here. If material should not appear at all, or a domain would rather not be probed by the census, that is a reasonable request and it will be honored — the census panel is a plain list in the repository and removing an entry takes one line.
The census identifies itself honestly on every request, with the user agent
28081999.me-census/2.0 and a link back to this site. It requests only
well-known files — robots.txt, llms.txt and
the like — which exist precisely to be read by machines. It never crawls site content.